422 M. BECQUEREL ON CHEMICAL DECOMPOSITION AND 
suppose that the tube contains it in excess: at first protoxide is pro- 
duced and crystallized ; the solution gradually loses its colour ; then be- 
comes colourless, and crystals of nitrate of ammonia are seen on the in- 
terior surface of the tube. The liquor now contains nothing but the 
saturated solution of this salt and some traces of copper. Sometimes it 
takes a year or more to obtain this last result, which depends on the 
quantity of deutoxide employed. All this is effected without any con- 
tact with the air, since the tube is hermetically sealed, and the forma- 
tion of the ammonia must have been owing to the hydrogen of the 
water and the azote of the nitric acid. 
When the quantity of the deutoxide is very small, the effect is as fol- 
lows : the protoxide crystals are formed equally on the plate of copper ; 
but by little and little, they lose some of their brilliancy, and experience 
at last a discoloration which stops at a certain point. The solution re- 
mains always coloured. The experiment is then terminated, and time 
produces no change in the solution. 
In order to explain the facts just mentioned, and to reascend to the 
cause of the electric phenomena by which they have been produced, 
we have found it necessary to analyse the octahedral crystals and the 
substance that replaces the deutoxide of the same metal. The change 
which the deutoxide undergoes is the only thing that can throw a light 
on the origin of the electric effects. 
Those crystals possess the following properties : their powder is red : 
it is soluble in ammonia or in hydrochloric acid without colouring either. 
The latter solution is made turbid by the water, and receives a blue tinge 
from ammonia. These characters indicate that the crystals really are 
protoxide of copper. 
Analysis of the Substance which replaces the Deutoxide of Copper 
in the Crystallization of the Protoxide. 
We took two grammes of this substance. After having well washed 
and dried it, we proceeded to operate on it immediately by means of car- 
bonate of potash. The filtrated liquor was gradually saturated with 
sulphuric acid, until there was no longer any alkaline reaction. Hav- 
ing now condensed the solution by evaporation and produced crystal- 
lization, we obtained 180 of nitrate of potash besides the mother-water 
which we neglected. 
The insoluble salt which remained on the strainer was carbonate of 
copper, which being dried and weighed amounted to 186. Now one 
gramme of nitrate of potash, if we admit that an atom of this salt includes 
two atoms of acid and one atom of base, will contain 085 of acid and 
0°45 of potash. 
In like manner the carbonate of copper, being formed of an atom of 
deutoxide of copper and an atom of carbonic acid, gives 1-2 of oxide 
and 0-4 of carbonic acid. 
ee 
